metadata
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_17_0
metrics:
- wer
model-index:
- name: whisper-small-id
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: mozilla-foundation/common_voice_17_0 id
type: mozilla-foundation/common_voice_17_0
config: id
split: None
args: id
metrics:
- name: Wer
type: wer
value: 0.16172444196946495
whisper-small-id
This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_17_0 id dataset. It achieves the following results on the evaluation set:
- Loss: 0.3506
- Wer: 0.1617
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0547 | 3.8462 | 1000 | 0.2570 | 0.1687 |
| 0.0034 | 7.6923 | 2000 | 0.3144 | 0.1589 |
| 0.0011 | 11.5385 | 3000 | 0.3358 | 0.1615 |
| 0.0007 | 15.3846 | 4000 | 0.3460 | 0.1614 |
| 0.0006 | 19.2308 | 5000 | 0.3506 | 0.1617 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1